Journal of Immunological Methods 336 (2008) 213–221
Contents lists available at ScienceDirect
Journal of Immunological Methods
j ourna l homepage: www.e lsev ie r.com/ locate / j im
Research paper
Seven week culture of functional human mast cells from buffycoat preparations
Mette Holm a,b,⁎, Hanne Busk Andersen b, Thea Eline Hetland b, Christine Dahl b,Hans Jürgen Hoffmann c, Steffen Junker a,1, Peter Oluf Schiøtz b,1
a Department of Human Genetics, University of Aarhus, DK-8000 Aarhus C, Denmarkb Department of Pediatrics, Aarhus University Hospital Skejby, DK-8200 Aarhus N, Denmarkc Department of Respiratory Diseases, Aarhus University Hospital, DK-8000 Aarhus C, Denmark
Article history:Received 14 March 2007Received in revised form 24 April 2008Accepted 25 April 2008Available online 27 May 2008
Functional, mature human mast cells have been generated by in vitro differentiation of CD133+/CD34+ progenitor cells isolated from e.g. cord blood, peripheral blood, bone marrow or fetalliver. However, the protocols published so far require long term cultivation, i.e. up to 15 weeksfor mast cell differentiation, which makes such approaches not only laborious but also costly.Here, we have developed a protocol for generating functional humanmast cells from peripheralblood already within 7 weeks. Human CD133+ progenitors were isolated from buffy coatpreparations of peripheral blood and cultured in the presence of stem cell factor (SCF) and IL-6for 7 weeks. IL-3 was added to the culture medium during the first 3 weeks, and fetal calf serum(FCS) added during the last week. In vitro differentiated CD133+ cells exhibited multiplecharacteristics of mature mast cells. Thus, cells contained tryptase and expressed functionallevels of FcεRI. Anti-IgE stimulation induced significant release of histamine and PGD2 and alsoof chemokines including MCP-1, IL-8, MIP-1α, and MIP-1β. The fact that our in vitrodifferentiated mast cells are derived from a generally available source of progenitor cellsmakes this novel protocol widely applicable to any patient group, irrespective of age. Moreover,this progenitor source is more readily available than e.g. bone marrow or cord blood-derivedprogenitors. Consequently, our protocol has great potential in studies on mast cell biology andmast cell pathology, and e.g. on evaluation of drug effects.
role in allergy is their expression of the high affinity Fc receptorfor IgE (FcεRI) (Kinet, 1999). This receptor becomes occupied byallergen-specific IgE produced by B cells. Aggregation of IgE-occupied FcεRI by antigen coordinates activation of intracellularsignalling pathways and results in degranulation of preformedmediators and synthesis and secretion of de novo synthesizedlipid mediators, cytokines, and chemokines (Galli et al., 2005).Mediators derived from mast cells such as histamine, leuko-trienes, andprostaglandinsare responsible for the IgE-dependentallergic reactions clinically recognized as anaphylactic reactions,acute asthma, and allergic rhinitis (Ishizaka et al., 1983).
Mast cells are derived fromCD34+pluripotent progenitor cellsthat aredefinedas c-kit (CD117)+, CD13+, andFcεRI−, and they lackT-cell (CD2), B-cell (CD19, CD20), macrophage (CD14), andeosinophil lineage surface markers (Kirshenbaum et al., 1991;Rodewald et al., 1996; Chen et al., 2005). In contrast to other
214 M. Holm et al. / Journal of Immunological Methods 336 (2008) 213–221
granulocytes that differentiate and mature within the bonemarrow, mast cells leave the bone marrow as progenitors andundergo their terminal differentiation in the tissue. Therefore,mature mast cells do not circulate in the peripheral blood. Thephenotype of mast cells varies in different anatomical compart-ments. These phenotypically distinct populations of humanmastcells are referred to as mast cell heterogeneity (Galli et al., 2005).
Primary human mast cells are difficult to isolate becauseonly limited amounts of cells can be retrieved from humantissues. Moreover, current methods are quite laborious(Schulman et al., 1982; Sellge and Bischoff, 2006). Instead,alternative means of obtaining human mast cells by in vitrodifferentiation have been successful using various sources ofprogenitor cells and variable culture conditions (Saito et al.,1996; Valent et al., 1992; Irani et al., 1992).
However, comparative studies of long term cultured mastcells (Iida et al., 2001) reveal significant quantitative as well asqualitative differences between mast cells depending on theirsource. Thus, yields between cord blood progenitor-derivedand adult progenitor-derived human mast cells differ greatly.Neonatal hematopoietic progenitor cells generate moredescendants than their adult counterparts. Functionally, theadult-type progenitor-derived mast cells are reportedlysuperior evaluated by their ability to release histamine uponactivation (Inomata et al., 2005). It appears, therefore, thatperipheral blood is a preferable source for many studies.
Previously published protocols are based on long termcultures (Dahl et al., 2002; Wang et al., 2006b; Yamaguchiet al., 2007), up to 15 weeks, which makes in vitrodifferentiation of mast cells extremely laborious, costly butalso vulnerable to contaminations. Thus, the potentialadvantages of reduced culture time are evident. The aim ofthis study was to reduce culture time for in vitro differentia-tion of human mast cells and thereby introduce a novelprotocol that is more generally applicable to progenitor celldonors of different ages and with different diseases withaberrant mast cell function.
Here, we report a sevenweek suspension culture method forgenerating human mast cells from in vitro differentiatedprogenitors isolated from peripheral blood. In the current studywe used CD133+ progenitor cells (Dahl et al., 2002), becauseCD133 defines a biologically distinct subset of haematopoieticstem cells that includes not only CD34+ (bright) cells, but also asubset of CD34− (dim) cells. Such CD133+ cells possess long termrepopulating potential and could be critical in human mast celldevelopment (Miraglia et al., 1997; Yin et al., 1997; Bauer et al.,2007). The in vitro differentiated cells were morphologically andfunctionally comparable to previously described cultured humanmast cells as they stained negative for CD14 and CD23, butpositive for CD117, CD203c and tryptase, and at variable levels forFcεRI. Moreover, the cells responded to FcεRI activation byreleasing histamine and PGD2. Finally, we demonstrate that IgEcross-linking also induces significant production of chemokinessuch as MCP-1, IL-8, MIP-1α, and MIP-1β.
2. Materials and methods
2.1. Cytokines and reagents
Ficoll–Paque cell separation solution was purchased fromGE Healthcare (Uppsala, Sweden). AC133 cell isolation kit and
LS+ cell separation columns were purchased from MiltenyiBiotech (Bergisch Gladbach, Germany). StemSpan culturemedium was from Stem Cell Technologies (Vancouver,Canada). Fetal calf serum (FCS) and penicillin/streptomycinwere purchased from GIBCO BRL (Grand Island, NY, USA).Human recombinant stem cell factor (rhSCF), human rhIL-6,and human rhIL-3 were purchased from R&D Systems(Abingdon, UK). Human rIL-4 was purchased from PeproTech(London, England). IgG1 antibodies against chymase andtryptasewere purchased from Chemicon (Temecula, CA, USA).IgG1 RPE conjugated monoclonal antibody (mAb) againstCD117 (clone 104D2), and its isotype control were fromSerotec (Kidlington, UK). CD23/FITC (clone MHM6) and CD14/RPE (clone R0864) and their isotypes were purchased fromDako Cytomation (Glostrup, Denmark). CD203c/PE (clone97A6) and its isotype were from Immunotech (Marseille,France). Myeloma IgE (supernatant from the B-cell line U266)was a kind gift from Lars K. Poulsen, National UniversityHospital, Allergy Unit. Anti-FcεRIα mAb, CRA-1 was pur-chased from Kyokuto Pharmaceuticals (Tokyo, Japan). Thealkaline phosphatase anti-alkaline phosphatase (APAAP) kitused was from Dako Cytomation (Glostrup, Denmark). DAPI,4′,6-diamidino-2-phenylindole was purchased from Sigma-Aldrich.
2.2. Cell preparation
Human mast cells were generated from CD133+ cells frombuffy coat preparations. With the acceptance from the localethical committee, fresh buffy coats were obtained from theblood bank of Aarhus University Hospital, Denmark. From atotal volume of 470 ml full blood the concentrated leukocytesuspension (buffy coat) was collected via Optipress II (BaxterHealthcare, Deerfield, IL., USA). The 50 ml buffy coat wasdiluted with triple volume PBS and gently layered on a Ficoll–Paque gradient before centrifugation at room temperature for30 min, 450 ×g. The interface layer of mononuclear cells washarvested and washed in PBS at 4 °C for 5 min, 450 ×g.
2.3. Purification of CD133+cells
CD133+ cells were separated using AC133 cell isolation kitand a magnetic LS+ separation column according to themanufacturer's instructions. Briefly, the mononuclear cellswere resuspended in 300 µl MACS buffer (PBS containing BSA,EDTA, pH 7.2) and pre-treated with FcR blocking reagent at100 µl/108 cells. Cells reacted with AC133 MicroBeads andincubated for 30 min at 4 °C. After incubation, the cells werewashed in MACS buffer. The pellet was resuspended in 500 µlMACS buffer and passed through the separation column. Thecolumn was washed four times with 3 ml MACS buffer in themagnetic field, and CD133+ cells were retained in the columnas the lineage negative cells were eluted. The column wasremoved from the magnet and the CD133+ cells werecollected and counted in Trypan blue.
2.4. Cell cultures
Purified CD133+ cells were suspended at 5×105 cells/ml inStemSpan medium supplemented with 100 ng/ml rhSCF,50 ng/ml rhIL-6, 1 ng/ml rhIL-3, 100 µg penicillin/
215M. Holm et al. / Journal of Immunological Methods 336 (2008) 213–221
streptomycin per ml, and grown for 3 weeks after which IL-3was omitted from the culture media. From week 6, 10% FCSwas added, and mature PBMC were analysed in week 7. Inorder to evaluate whether culture time had implications forthe phenotype and function of the cells, n=6 batches of cellswere cultured for 11 weeks and analysed in weeks 7, 9 and 11.
Cells were cultured in suspension for the entire period.Medium was totally renewed weekly. Cells were spun at 4 °Cfor 5 min, 450 ×g, the culture medium gently aspirated andreplaced by fresh medium. Weekly, cell cultures were stainedwith Trypan blue (0.5% in saline, Pharmacy, UniversityHospital of Aarhus, Denmark) to assess cell viability andwith Alcian blue to assess the purity and numbers of mastcells by metachromasi (Gilbert and Ornstein, 1975).
Studies on the influence of cell density on mast cellphenotype and survival were carried out (data not shown).Four batches of CD133+ cells were split and cultured at threedifferent cell densities: 5×105 cells/ml, 7.5×105 cells/ml, and1×106 cells/ml. With increasing cell densities, cell phenotypewas not affected with respect to surface receptor expression,protease content, total histamine content or histaminerelease, but the yield of mature cells decreased. Thus, in thepresent study progenitor cells were suspended in culture
Fig. 1. CD133+ peripheral blood-derived mast cell, 7 weeks. A and B. Human peripherfor 6 weeks and with 10% FCS for 1 week. Stem cell factor, IL-6, and IL-3 was added twith May–Grünwald–Giemsa show contents of cytoplasmic granules. C. Cell surfaceisotype control. Cells were counterstained with DAPI, 4′,6-diamidino-2-phenylindo
medium at 5×105 cells/ml and this cell density was main-tained during the entire culture period.
2.5. Immunocytochemical staining
Immunostaining for intracellular stores of tryptase wasperformed using the alkaline phosphatase anti-alkalinephosphatase (APAAP) method described by Craig et al.(1986). Cytospins were prepared by cytocentrifugation(250 ×g) using a Shandon Cytospin 3. The cytospins wereair-dried, fixed in methanol/acetone solution, and wereincubated overnight in a humidified chamber at 4 °C withmAb against tryptase. After washing with Tris buffered saline(TBS), the cytospins were incubated with rabbit anti-mouseIgG for 30 min at room temperature. After washing, thecytospins were incubated with the alkaline phosphatasemouse anti-alkaline phosphatase for 30 min. Negativecontrols were performed using isotype-matched mouseIgG1 instead of the primary antibody and by the T-cell lineJurkat.
Cell surface expression of FcεRI α-chain was assessed byboth flow cytometry and immunostaining as this receptor isessential for the use of cell cultures in allergen-mediated
al blood-derived mast cells (PBMC) cultured in serum-free StemSpanmediumo the media but IL-3 was omitted after the first week of culture. Cells stainedexpression of FcεRI α-chain by immunostaining with CRA-1 (green) or D. Anle, which is a fluorescent stain that binds strongly to nuclear DNA (blue).
Table 1Surface receptor expression of the high affinity IgE receptor (FcεRI), CD117,and CD203c on PBMC
CD133+ PBMC were cultured for 7 weeks with SCF, IL-6 and initially with IL-3.Cells preincubated for 24 h with myeloma IgE (2 µg/ml) before incubation withanti-FcεRIα (CRA-1) or an isotype control for 30min,washing, and labellingwithFITC-conjugated anti-IgG2b antibody. Expression of CD117/RPE, and CD203c/PEwas assessed in unstimulated cells by primary conjugated antibodies comparedto an isotype control. Flow cytometry was performed using a FACS Calibur.Data are presented as mean±SEM of MFI (mean fluorescence intensity) andpercentage of positive cells calculated by Overton Subtraction.
216 M. Holm et al. / Journal of Immunological Methods 336 (2008) 213–221
studies. Unsensitized cells were incubatedwith CRA-1 (0.1 µg)primary monoclonal antibody against the high affinity IgEreceptor for 30 min at 4 °C and washed with PBS containing0.1% HSA followed by incubation for 30 min at 4 °C withsecondary FITC labelled antibody. Isotype control wasperformed using an isotype-matched IgG2b instead of CRA-1. Cells were counterstained with DAPI, 4′,6-diamidino-2-phenylindole (Fig. 1C and D).
2.6. May–Grünwald–Giemsa staining
Cytospins were prepared by cytocentrifugation (250 ×g)using a Shandon Cytospin 3 and air-dried. They wereincubated in May–Grünwald solution (Merck 1424, Darm-stadt, Germany) for 4 min and washed in PBS (pH 6.4) for4min. After rinsing inwater cytospins were stained in Giemsa(Merck 9204) diluted 1/10 in PBS, pH 6.4, for 8 min and rinsedin water. Cytospins air-dried and were seen under a light-microscope (Olympus, BX40 (Fig. 1A and B).
2.7. Flowcytometry
Surface expression of CD117, FcεRI, CD14, CD23, andCD203c was assessed by flow cytometry using a FACS Calibur(BD, San José, Ca, USA). Human cultured mast cells weresensitized with human myeloma IgE (2 µg/ml) for 24 h priorto the analysis. Cells were washed twice in PBS containing0.1% HSA before labelling with antibody in dilution 1/10. Cellswere incubated for 30 min at 4 °C with conjugatedmonoclonal receptor antibodies. The CRA-1 labelled cells,after the primary antibody incubation, were again incubatedat 4 °C for 30 min with secondary FITC-conjugated anti-mouse IgG2b antibody after twowashes in PBS with 0.1% HSA.Cell viability was assessed by exclusion of propidium iodide.Data were analysed with FCS-express. Overton Subtractionwas used to calculate the net percentage of positive cells(Overton, 1988).
2.8. Mediator release from in vitro differentiated human mastcells
The presence of mast cells was assessed by assays forrelease of substantial amounts of histamine and PGD2 inresponse to cross-linking of the FcεRI. PGD2 is considered acharacteristic marker for mast cells, whereas histamine is alsoreleased by basophils in less quantity.
Mature mast cells were incubated with human myelomaIgE (2 µg/ml) for 24 h at 37 °C in StemSpan medium suppliedwith 100 ng/ml rhSCF and 50 ng/ml rhIL-6. The sensitizedmast cells were washed and 104 cells were resuspended (105
cells/ml) in 100 µl pipes buffer (10 mM Pipes, 150 mM Na-acetate, 5 mMK-acetate, 0.6 mM CaCl2, 1.1 mMMgCl2, glucose1 mg/ml, human serum albumine 0.3 mg/ml, and heparin15 IE/ml, pH 7.4) with rhSCF and rhIL-6. Cells were activatedby incubation with 100 µl anti-IgE (Dako, Denmark, 5 µg/ml)for 30 min at 37 °C. Histamine release was stopped by theaddition of 100 µl ice-cold buffer pipes buffer (10 mM Pipes,150 mM Na-acetate, 5 mM K-acetate, 0.6 mM CaCl2, 1.1 mMMgCl2, pH 7.4) followed immediately by centrifugation(450 ×g) for 5 min at 4 °C. The cell pellets and supernatants(150 µl) were collected separately for the histamine assay. For
determination of histamine release, a microfiber-basedmethod was used (Skov et al., 1985). The method is basedon chemical extraction of histamine to ground glass micro-fibers and measures histamine by fluorescence of a hista-mine-o-phtaldehyde complex (Skov et al., 1985).
For the quantitation of total histamine 105 unstimulatedmast cells were washed in PBS at 500 ×g for 5 min at 4 °C andresuspended in 100 µl PBS. To stabilize histamine perchloricacid was added (40 µl, 7%), and incubated at roomtemperature for 5 min before the solution was neutralizedby the addition of 400 µl pipes buffer. The intracellularhistamine content is determined by the above describedmethod (Skov et al., 1985). The sensitivity of the assay is0.63 ng histamine, 2 HCl.
Measurement of prostaglandin D2 (PGD2) from anti-IgEstimulatedmast cells wasmeasured using a PGD2-MOX-EIA kit(Cayman Chemicals, Ann Arbour, MI) according to manufac-turer's instructions. The sensitivity of the assaywas6pg/ml. Theintra-assay coefficient of variation was less than 10%.
2.9. Analysis of chemokine secretion from in vitro differentiatedhuman mast cells
Cells were sensitized for 3–5 days with rhIL-4, 10 ng/ml,prior to analysis of chemokine secretion. Mediawere renewedbefore stimulation in order to quantify constitutive as well asactivation dependent chemokine secretion from human mastcells. Cells were incubated 24 h with human myeloma IgE(2 µg/ml) at 37 °C before the FcεRI were cross linked for aperiod of 2 h, 4 h, 8 h, or 24 h. The spent media were collectedby centrifugation and accumulated chemokines were quanti-fied using the Luminex® technology (Langouche et al., 2005)and a Bio-Plex® human cytokine assay (Bio-Rad, CA, USA) forsimultaneous quantitation. A Bio-Rad cytokine assay (Breueland De Ponti, 2006) was used for determination of humanchemokines in media collected from stimulated mast cells.Beads were read on the Bio-Plex® suspension array system.Samples were read in duplicate. Sensitivity was 2 pg/ml andthe intra-assay coefficient of variation varied from 4–13%.
2.10. Statistics
All results are expressed as mean±standard error of themean values for n independent batches of mast cell cultures.
217M. Holm et al. / Journal of Immunological Methods 336 (2008) 213–221
Basic phenotypic and functional analyses such as immunostain-ing, surface receptor expression, and histamine release wereperformedonn=57batchesof cell cultures. Thesedata confirmeda high degree of reproducibility of the independent batches.Throughout this manuscript n=57 unless otherwise indicated. Inaddition, we present data from n=9 batches that have beenextensively analysed for multiple phenotypic markers as well asfor kinetics of chemokine gene activation. These batches arerepresentative for the cells generated by our novel protocol.
Fig. 2. IgE-dependent histamine release and secretion of PGD2 from PBMC. A. CD133before IgE receptor cross-linking for 30 min. with anti-IgE in increasing concentratirelease was determined by a microfiber-based method and illustrated as % histaminvitro differentiated for 7, 9 or 11 weeks. Anti-IgE induced histamine release in % of tillustrated relatively to histamine release from cells cultured for 7 weeks (left y-axis).6.7±0.8% inweek 9; 9.7±2% inweek 11. IgE-dependent secretion of PGD2 wasmeasurcalculated in ng/ml, and expressed relatively to PGD2 secretion from cells cultured foalone was 5.5±2% in week 7; 2.1±0.6% in week 9; 4.5±1.5% in week 11.
3. Results
3.1. Phenotypic analysis of in vitro differentiated human mastcells
Following Ficoll–Paque sedimentation and subsequentpositive selection with antibodies specific for the AC133antigenwe harvested 2.5±0.2×106 CD133+ progenitors from abuffy coat preparation. Preliminary results showed an optimal
+ PBMC cultured for 7 weeks were preincubated with myeloma IgE (2 µg/ml)ons ((µg/ml, DAKO). IgE stimulated cells served as controls (n=6). Histaminee released of total histamine per cell. B. Functional analyses of CD133+ cells inotal histamine in the cells, and histamine release from 9 to 11 weeks cells isSpontaneous histamine release induced by IgE alonewas 7.7±2.2% inweek 7;ed by PGD2-MOX-EIA kit (Cayman Chemicals). PGD2 secretion from 104 cells isr 7 weeks (right y-axis), n=6. Spontaneous secretion of PGD2 induced by IgE
218 M. Holm et al. / Journal of Immunological Methods 336 (2008) 213–221
culture period of 7 weeks and these cells were thereforeextensively analysed for various surface markers and func-tionality. However, to elucidate mast cell phenotype afterprolonged culture period cells were also analysed in weeks 9and 11 for expression of FcεRI and for receptor-mediatedrelease of histamine and PGD2 (n=6). The cells were in vitrodifferentiated with 100 ng/ml SCF and 50 ng/ml IL-6 asdescribed in the Materials and methods section. During thefinal week, i.e. week 7, 9 or 11, FCS (10% v/v) was added to theculture media. After 7 weeks were generated 8.2±1.2×106
cells (Fig. 1). Following metachromatic staining with Alcianblue 99.4±0.3% of the cells were positive in this simple andindicative test for a mast cell phenotypic marker.
The protease tryptase is a major mast cell characteristic. Byimmunostaining, 80.0±5.4% of the cells were tryptasepositive after 7 weeks of culture and no significant changeswere observed in weeks 9 and 11. High intracellular stores ofhistamine is another mast cell characteristics. The mean totalhistamine content of the peripheral blood-derived mast cellswas 15.5±5.3 pg/cell.
The sensitized cells were also analysed by FACS forexpression of a broad range of surface markers attributed todifferent lineages in the hematopoietic system. At 7 weeks ofculture 88.3±2.2% of the cells were positive for CD117, 31.4±3.3% of the cells were FcεRIα positive (Table 1) and theactivation marker CD203c was expressed on 72.6±3.2% of thecells. The density on the cells of the same receptors (meanfluorescent intensity or MFI) was 649.2±73.6 for CD117, 77.4±7.8 for FcεRIα, and 215.6±19.0 for CD203c, respectively. Thecells were CD14 and CD23 negative, demonstrating that theyare not monocytic. Collectively, these analyses indicate thatthe cultures are of mast cell lineage.
Six of the 57 cell batches were propagated for a total of11weeks, and the FcεRI expressionwas assessed on a single-cell
Fig. 3. PBMC release chemokines upon anti-IgE stimulation. CD133+ PBMC cultured fowith myeloma IgE (2 µg/ml) before addition of anti-FcεRIα (CRA-1) for indicated timcentrifugation of the cells and accumulated chemokines quantified using the Lumquantitation. Beads were read on the Bio-Plex® suspension array system. Samples0.3 ng IL-8/ml, 0.7±0.2 ng MCP-1/ml, 0.03±0.01 ng MIP-1α/ml, and 0.2±0.05 ng M
basis using immunofluorescence microscopy (Fig. 1C and D).Using this approach, 50.8±9.2% of the cells were FcεRI positivein week 7, declining to 36.7±10.6% in week 11. Staining wasneither visible onmast cells stainedwith an isotype (Fig.1C) noron human Jurkat T cells stained with the FcεRI antibody. Theapparent discrepancy between the estimated numbers of FcεRIpositive cells in the FACS and the microscopical analyses inweek7maybe explainedby the rathermodest separationof thehistograms generated from the isotype and FcεRI specificantibodies in the FACS analysis, which results in an under-estimation of the size of the positive cell population. In fact, theproportion of FcεRI positive cells scored in microscopicalanalyses has consistently been larger than that obtained inthe FACS analyses.
3.2. Mediator release from activated human mast cells
Mast cell functionality was assessed by quantification ofhistamine released from cell secretory granules into themedium, and concomitantly by generation of the significantmast cell arachidonic acid product, prostaglandin D2 (PGD2)after 7, 9, or 11 weeks of differentiation. Cells cultured for7 weeks were incubated with 2 °µg/ml human myeloma IgEfor 24 h followed by anti-IgE stimulation for 30 min. Cellssensitized with IgE released 1.3±0.1% of the total histaminecontent whereas subsequent anti-IgE stimulation increasedhistamine release to 53.5±3.0% corresponding to the releaseof 8.9±1.4 pg histamine/cell. Thus, the cells release substan-tial amounts of their total histamine content within the rangeof that previously published for mast cells (Kinoshita et al.,1999; Kikuchi et al., 2002; Lappalainen et al., 2007). Mast cellfunctionality assessed by anti-IgE mediated histamine releasewas not altered by prolonged culture time (Fig. 2B, left y-axis).The IgE induced leak of PGD2 was low (0.3±0.1 ng PGD2/ml
r 7 weeks were preincubated for 3–5 days with rhIL-4 (10 ng/ml) and for 24 hes of 0, 2 h, 4 h, 8 h, or 24 h (106 cells/ml). Culture media were collected afterinex® technology and a Bio-Plex® human cytokine assay for simultaneouswere read in duplicate. Spontaneous release of chemokines at 0 h was: 0.5±IP-1β/ml.
219M. Holm et al. / Journal of Immunological Methods 336 (2008) 213–221
from 104 cells), and the spontaneous release was even belowthis level (data not shown). However, anti-IgE stimulationinduced secretion of 8.1±0.7 ng PGD2/ml from 104 cells, andthe PGD2 secretion increased significantly from cells in vitrodifferentiated for 9 and 11 weeks (Fig. 2B right y-axis,pb0.05). This suggests an optimal phenotype of the mastcells in weeks 7 to 9.
3.3. Effect of mast cell activation on chemokine release
Mast cells are effector cells contributing to the inflamma-tory reaction through directly released mediators and wide-spread communication with other inflammatory cells. Theexperiments on chemokine secretion from mast cells arefocused on 7 week cultures of cells.
We have previously shown that in vitro differentiated cordblood-derived mast cells produce chemokines in response toFcεRI activation (Holm et al., 2006). Here we assessedchemokine production in untreated cultures or followingFcεRI activation for up to 24 h, the kinetics of which isdisplayed in Fig. 3. Whereas constitutive chemokine produc-tion was very low, significant quantities accumulated duringactivation. Thus, e.g. at 8 h after FcεRI cross-linking in vitrodifferentiated peripheral blood-derived mast cells (106/ml)produced: 2.1±0.4 ng MCP-1/106 cells, 21.4±3.1 ng IL-8/106
cells, 27.1±2.3 ng MIP-1α/106 cells, and 18.3±2.3 ng MIP-1β/106 cells (n=7).
Collectively, these data show that the in vitro differen-tiated cells derived from buffy coat preparations are func-tionally intact mast cells in a 7 to 9 week window but alreadyfrom week 7 the cells are highly responding to activation viathe high affinity receptor for IgE by releasing mediatorsinvolved in chemotaxis and activation of inflammatory cells inan inflammatory reaction.
4. Discussion
We have efficiently generated functional humanmast cellswithin 7 weeks of in vitro differentiation of CD133+ precursorsisolated from buffy coat preparations. Expression of mast cellspecific receptors, secretion of histamine and PGD2 inresponse to FcεRI cross-linkage, and expression of tryptaseall confirmed that these cells are mast cells. Previouslypublished in vitro differentiation protocols (Wang et al.,2006b; Valent et al., 1992; Dahl et al., 2002) require longterm cultures, up to 15 weeks, which makes cultivation ofmast cells not only very laborious but also expensive andextremely vulnerable to infections. Advantages of our novelmethod are a) greatly reduced culture periods and b) rathereasy accessibility to donors. Here, we also simplified existentprotocols (Wang et al., 2006b; Iida et al., 2001; Saito et al.,2006) by maintaining the cells in suspension during theentire culture period. However, the most remarkable advan-tage of the generated mast cells is their high level ofimmunological functionality compared to the cells generatedby recently published protocols (Lappalainen et al., 2007;Yamaguchi et al., 2007). Lappalainen et al. (2007) report ananti-IgE mediated histamine release of about 35% (n=6) oftotal histamine compared to 53% (n=57) in this study. Thisillustrates the impact of culture conditions on mast cellphenotype and the various factors influencing the outcome of
cells (i.e. culture time, serum-supplemented cultures) arecommented on in this discussion.
Mature mast cells are generated in only 7 weeks by thisprotocol. However, to address whether mast cell phenotypeor functionality alters with increasing culture time, we furtheranalysed the cells after 9 and 11 weeks of differentiation(n=6). We found an optimal mast cell phenotype after 7 to9 weeks of in vitro differentiation as FcεRI expressiondecreased with prolonged culture period. However, thealtered FcεRI expression had no impact on mast cellfunctionality assessed by histamine release suggesting a lowreceptor threshold for induction of an anti-IgE mediatedresponse in mast cells. This is further illustrated by the rapiddose response in anti-IgE mediated histamine release shownin Fig. 2A.
Previously,we (Dahl et al., 2002) and others (Kinoshita et al.,1999; Ahn et al., 2000) have shown that human mast cellsproliferate in vitro for a longer period of time in serum-freeconditions than in fetal calf serum-supplemented conditions,but for maximal differentiation of in vitro generated mast cellsFCS is required. Therefore, the addition of FCS and the selectionof anoptimal timepoint for the addition are crucial for reducingculture time but yet maintaining an adequate yield of highlyfunctionalmaturemast cells. Lappalainen et al. (2007) present aprotocol using serum-free culture conditions.
In the present study a final cell number of 8.2±1.2×106
mature mast cells were generated from one buffy coatpreparation of 50 ml without cytokine priming of the donor.This is comparablewith the yield of cells reported by Iida et al.(2001) from 12–14 weeks cultures of peripheral bloodprogenitor cells, but inferior to the yield reported in morerecent protocols (Wang et al., 2006b; Lappalainen et al., 2007)where 2.0±1.0×107 mast cells were cultured in 9 to 12 weeks.In our study prolonged culture time had no significant effecton the yield. Thus, the yield of mature mast cells is dependenton differences in culture conditions other than culture time.Characteristically, mast cells contain histamine which isreleased by degranulation. The mean intracellular histaminelevel in the cultured cells was 15.5±5.3 pg/cell. Comparablelevels of total histamine (15.5±3 pg/cell) were reported inmast cells also cultured with SCF 100 ng/ml and IL-6 50 ng/ml(Kinoshita et al., 1999).
Sensitized, resting cells released low levels of the arachi-donic acid product PGD2 with significant induction followingcross-linkage of FcεRI (5 µg/ml, 8.1±0.7 ng PGD2/ml/104 cells).Compared hereto long term cultured peripheral blood-derived mast cells are reported to secrete PGD2 in a nine-fold reduced level upon activation (Wang et al., 2006a). Thisreduced level of PGD2 secretion is within the range of thatpreviously found in cord blood-derivedmast cells (2.4±0.3 ngPGD2/104 cells, n=14, unpublished), (Obata et al., 1996). Thus,short term culture generates functionally intact human mastcells with a remarkable capability of PGD2 secretion. Thisfunctional difference may be related to decreased cell surfaceexpression of FcεRI on long term differentiated mast cells butno data on receptor expression was reported.
Tryptase is the most specific mast cell marker (Som-merhoff, 2001). Whereas some mast cells express onlytryptase (MCT), others express both chymase and tryptase(MCTC). Immunocytochemical staining studies showed thatthe peripheral blood-derived mast cells phenotypically
220 M. Holm et al. / Journal of Immunological Methods 336 (2008) 213–221
resemble the subtype of mast cells referred to as MCT thatpredominate in the human alveolar wall, the epithelium ofthe lung, and small intestinal mucosa (Forsythe and Ennis,2000).
Following mast cell activation via the FcεRI, the tran-scriptional levels of several CC chemokines are markedlyincreased (Nakajima et al., 2002; Gonzalez-Espinosa et al.,2003). In accordance with these studies we previouslydemonstrated a significant augmentation in the accumula-tion of the chemokines MCP-1, MIP-1β, and IL-8 from cordblood-derived mast cells (Holm et al., 2006). Using our novelprotocol we now demonstrate that mast cells generated frombuffy coat progenitors also secrete IL-8, MIP-1α, and MIP-1βbut in quantities far exceeding those from CBMC (Holm et al.,2006).
The amplitudes of FcεRI-mediated IL-8 and MIP-1αproduction from mast cells sensitized with IgE and IL-4 differfrom that reported on 12 week cultures of PBMC activatedunder similar conditions (Wakahara et al., 2001). Thus,culture conditions appear to influence the level of chemokineproduction from PBMC but, importantly, the kinetics of FcεRI-mediated chemokine production is comparable in the studies(Wakahara et al., 2001). Kinetics of MCP-1 and MIP-1β areparallel as maximal accumulation of these chemokines occursat 4 h of activation after which the accumulation reaches asteady-state level. To our knowledge, this is the first report onFcεRI-mediated kinetics of MCP-1 and MIP-1β in peripheralblood-derived mast cells.
In conclusion, we have developed a method to in vitrodifferentiate human mast cells derived from CD133+ progeni-tors in buffy coat preparations, in a manner that considerablyreduce time and consequently reduce labour time, andtherefore also the risk of culture infections. This short termculture protocol will be useful for the evaluation of mast cellbiology and effect of drugs on these cells. Moreover, as themethod only requires an easily accessible blood sample itprovides the possibility to study mast cells from individualmast cell populations and mast cell pathology in differentgroups of patients.
Acknowledgement
We thank technician Ellen Margrethe Raaby for herpersistent work in cell culturing and Christian Knudsen forhelping with photographic documentation. Lars K. Poulsen,Professor, PhD, National University Hospital, Allergy Unitkindly suppliedwithmyeloma IgE. Troels K. Hansen, MD, PhD,Department of Endocrinology, Aarhus University Hospitalkindly contributed to analysis of cytokine production by usingthe Luminex® technology.
This work was supported by ALK-Abelló, LundbeckFoundation, Denmark, Danish Medical Research Council, theAarhus University Foundation, Denmark, Glaxo, Smith andKline, England, the Danish Asthma and Allergy Association,The Velux Foundation, the Danish Lung Association, theFoundation of 1870, Denmark, the Institute for ExperimentalClinical Research, Denmark, Dagmar Marshall Foundation,Fund for Medical Science Research, Denmark, Helga and PeterKornings Foundation, King Christian X.'s Foundation, theNOVO Nordic Foundation, and the Danish Pediatric AsthmaCenter.
References
Ahn, K., Takai, S., Pawankar, R., Kuramasu, A., Ohtsu, H., Kempuraj, D., Tomita,H., Iida, M., Matsumoto, K., Akasawa, A., Miyazaki, M., Saito, H., 2000.Regulation of chymase production in human mast cell progenitors. J.Allergy Clin. Immunol. 2, 321.
Bauer, N., Fonseca, A.V., Florek, M., Freund, D., Jaszai, J., Bornhauser, M.,Fargeas, C.A., Corbeil, D., 2007. New insights into the cell biology ofhematopoietic progenitors by studying prominin-1 (CD133). CellsTissues Organs.
Breuel, K.F., De Ponti, W.K., 2006. Measurement of mast cell cytokine releaseby multiplex assay. Methods Mol. Biol. 315, 217.
Chen, C.C., Grimbaldeston, M.A., Tsai, M., Weissman, I.L., Galli, S.J., 2005.Identification of mast cell progenitors in adult mice. Proc. Natl. Acad. Sci.U. S. A 32, 11408.
Craig, S.S., DeBlois, G., Schwartz, L.B., 1986. Mast cells in human keloid, smallintestine, and lung by an immunoperoxidase technique using a murinemonoclonal antibody against tryptase. Am. J. Pathol. 3, 427.
Dahl, C., Saito, H., Nielsen, H.V., Schiotz, P.O., 2002. The establishment of acombined serum-free and serum-supplemented culture method ofobtaining functional cord blood-derived human mast cells. J. Immunol.Methods 1–2, 137.
Forsythe, P., Ennis, M., 2000. Clinical consequences of mast cell heterogeneity.Inflamm. Res. 4, 147.
Galli, S.J., Kalesnikoff, J., Grimbaldeston, M.A., Piliponsky, A.M.,Williams, C.M.,Tsai, M., 2005. Mast cells as “tunable” effector and immunoregulatorycells: recent advances. Annu. Rev. Immunol. 23, 749.
Gilbert, H.S., Ornstein, L., 1975. Basophil counting with a new stainingmethod using alcian blue. Blood 2, 279.
Gonzalez-Espinosa, C., Odom, S., Olivera, A., Hobson, J.P., Martinez, M.E.,Oliveira-Dos-Santos, A., Barra, L., Spiegel, S., Penninger, J.M., Rivera, J.,2003. Preferential signaling and induction of allergy-promoting lym-phokines uponweak stimulation of the high affinity IgE receptor onmastcells. J. Exp. Med. 11, 1453.
Holm, M., Kvistgaard, H., Dahl, C., Andersen, H.B., Hansen, T.K., Schiotz, P.O.,Junker, S., 2006.Modulation of chemokine gene expression in CD133 cordblood-derived human mast cells by cyclosporin A and dexamethasone.Scand. J. Immunol. 5, 571.
Iida, M., Matsumoto, K., Tomita, H., Nakajima, T., Akasawa, A., Ohtani, N.Y.,Yoshida, N.L., Matsui, K., Nakada, A., Sugita, Y., Shimizu, Y., Wakahara, S.,Nakao, T., Fujii, Y., Ra, C., Saito, H., 2001. Selective down-regulation ofhigh-affinity IgE receptor (FcepsilonRI) alpha-chain messenger RNAamong transcriptome in cord blood-derived versus adult peripheralblood-derived cultured human mast cells. Blood 4, 1016.
Inomata, N., Tomita, H., Ikezawa, Z., Saito, H., 2005. Differential geneexpression profile between cord blood progenitor-derived and adultprogenitor-derived human mast cells. Immunol. Lett. 2, 265.
Irani, A.M., Nilsson, G., Miettinen, U., Craig, S.S., Ashman, L.K., Ishizaka, T.,Zsebo, K.M., Schwartz, L.B., 1992. Recombinant human stem cell factorstimulates differentiation of mast cells from dispersed human fetal livercells. Blood 12, 3009.
Ishizaka, T., Conrad, D.H., Schulman, E.S., Sterk, A.R., Ishizaka, K., 1983.Biochemical analysis of initial triggering events of IgE-mediatedhistamine release from human lung mast cells. J. Immunol. 5, 2357.
Kinet, J.P., 1999. The high-affinity IgE receptor (Fc epsilon RI): fromphysiology to pathology. Annu. Rev. Immunol. 17, 931.
Kinoshita, T., Sawai, N., Hidaka, E., Yamashita, T., Koike, K., 1999. Interleukin-6directly modulates stem cell factor-dependent development of humanmast cells derived from CD34(+) cord blood cells. Blood 2, 496.
Kirshenbaum, A.S., Kessler, S.W., Goff, J.P., Metcalfe, D.D., 1991. Demonstrationof the origin of human mast cells from CD34+ bone marrow progenitorcells. J. Immunol. 5, 1410.
Langouche, L., Vanhorebeek, I., Vlasselaers, D., Vander, P.S., Wouters, P.J.,Skogstrand, K., Hansen, T.K., Van den, B.G., 2005. Intensive insulin therapyprotects the endothelium of critically ill patients. J. Clin. Invest 8, 2277.
Lappalainen, J., Lindstedt, K.A., Kovanen, P.T., 2007. A protocol for generatinghigh numbers ofmature and functional humanmast cells fromperipheralblood. Clin. Exp. Allergy 9, 1404.
Miraglia, S., Godfrey,W., Yin, A.H., Atkins, K.,Warnke, R., Holden, J.T., Bray, R.A.,Waller, E.K., Buck, D.W.,1997. A novel five-transmembrane hematopoieticstem cell antigen: isolation, characterization, and molecular cloning.Blood 12, 5013.
Nakajima, T., Inagaki, N., Tanaka, H., Tanaka, A., Yoshikawa, M., Tamari, M.,Hasegawa, K., Matsumoto, K., Tachimoto, H., Ebisawa, M., Tsujimoto, G.,Matsuda, H., Nagai, H., Saito, H., 2002. Marked increase in CC chemokinegene expression in both human and mouse mast cell transcriptomes
221M. Holm et al. / Journal of Immunological Methods 336 (2008) 213–221
following Fcepsilon receptor I cross-linking: an interspecies comparison.Blood 12, 3861.
Obata, T., Nagakura, T., Kanbe, M., Masaki, T., Maekawa, K., Yamashita, K.,1996. IgE-anti-IgE-induced prostaglandin D2 release from culturedhuman mast cells. Biochem. Biophys. Res. Commun. 3, 1015.
Rodewald, H.R., Dessing, M., Dvorak, A.M., Galli, S.J., 1996. Identification of acommitted precursor for the mast cell lineage. Science 5250, 818.
Saito, H., Ebisawa, M., Tachimoto, H., Shichijo, M., Fukagawa, K., Matsumoto,K., Iikura, Y., Awaji, T., Tsujimoto, G., Yanagida, M., Uzumaki, H., Takahashi,G., Tsuji, K., Nakahata, T., 1996. Selective growth of human mast cellsinduced by Steel factor, IL-6, and prostaglandin E2 from cord bloodmononuclear cells. J. Immunol. 1, 343.
Saito, H., Kato, A., Matsumoto, K., Okayama, Y., 2006. Culture of human mastcells from peripheral blood progenitors. Nat. Protoc. 4, 2178.
Sommerhoff, C.P., 2001. Mast cell tryptases and airway remodeling. Am. J.Respir. Crit Care Med. 10 Pt 2, S52.
Valent, P., Spanblochl, E., Sperr, W.R., Sillaber, C., Zsebo, K.M., Agis, H., Strobl,H., Geissler, K., Bettelheim, P., Lechner, K., 1992. Induction of differentia-tion of human mast cells from bone marrow and peripheral bloodmononuclear cells by recombinant human stem cell factor/kit-ligand inlong-term culture. Blood 9, 2237.
Wakahara, S., Fujii, Y., Nakao, T., Tsuritani, K., Hara, T., Saito, H., Ra, C., 2001.Gene expression profiles for Fc epsilon RI, cytokines and chemokinesupon Fc epsilon RI activation in human cultured mast cells derived fromperipheral blood. Cytokine 4, 143.
Wang, X.S., Yip, K.H., Sam, S.W., Lau, H.Y., 2006b. Buffy coat preparation is aconvenient source of progenitors for culturing mature human mast cells.J. Immunol. Methods 1–2, 69.
Yamaguchi, M., Azuma, H., Fujihara, M., Hamada, H., Ikeda, H., 2007.Generation of a considerable number of functional mast cells with ahigh basal level of FcepsilonRI expression from cord blood CD34+ cells byco-culturing themwith bone marrow stromal cell line under serum-freeconditions. Scand. J Immunol 6, 581.
Yin, A.H., Miraglia, S., Zanjani, E.D., Almeida-Porada, G., Ogawa, M., Leary, A.G.,Olweus, J., Kearney, J., Buck, D.W., 1997. AC133, a novel marker for humanhematopoietic stem and progenitor cells. Blood 12, 5002.
